[Shake-dev] Re: shakemap error
Bruce Worden
cbworden at usgs.gov
Thu Mar 8 23:34:51 GMT 2007
Hi Zehra,
At its simplest, a fault file is just pairs of points separated by
spaces or commas:
latitude1, longitude1
latitude2, longitude2
...
or
latitude1 longitude1
latitude2 longitude2
...
A third value may be added to each line for depth.
For the dist_rjb() function (i.e., Joyner-Boore distance) each point
is considered to be connected to the next one in the file. Depth is
ignored as JB projects the points to the surface anyway. If the
first and last point are the same, the fault is treated as a polygon
(i.e., the fault plane projected to the surface), with all points
inside the polygon at zero distance.
dist_hypo() just computes the hypocentral distance (i.e., the
distance to a (subsurface) point only).
dist_rrupt() computes the distance to a multi-planar 3-D fault
surface. It is complicated to explain, though; see the Distance.pm
module for this function's documentation and fault definition syntax.
If you need some other distance measure, the functions in Distance.pm
should provide you with helpful examples.
There is currently no function that allows for separate, disconnected
fault segments, though there really should be. It wouldn't be that
hard to modify Fault.pm and dist_rjb() (in Distance.pm) to handle
that situation. I guess nobody has felt the need for it yet.
I home this helps.
Bruce
On Mar 6, 2007, at 12:25 AM, zehra cagnan wrote:
> Dear Dr Worden,
> I started to prepare Shakemap input files for past major
> earthquakes in Turkey.Especially I am interested with obtaining the
> instrumental intensity distribution for the Kocaeli earthquake. For
> cases like Kocaeli earthquake I will have to include a _fault.txt
> file. I am not clear about this file's format and whether I can
> include more than one segment for strike slip type fault mechanisms
> or not. I was wondering if you could send me an example _fault.txt
> fõle for this purpose.
> Thank you very much for all your help.
> Best Regards,
> Zehra Cagnan
>
>
> On 2/16/07, Bruce Worden <cbworden at usgs.gov > wrote:
> Hi Zehra,
>
> I'm glad to hear you got it working -- most of the issues with GMT
> are a result of the differences in the way the headers behave
> between Intel and Sun hardware. You may have hit on a combination
> of GMT and ShakeMap that just happens to work. I would recommend
> upgrading to ShakeMap V3.2, but that can wait until the official
> version is released (hopefully next week -- the code is done, I'm
> just waiting for a new distribution site).
>
> Bruce
>
> On Feb 16, 2007, at 2:00 PM, zehra cagnan wrote:
>
>> Hi Bruce
>> Thanks a lot for all your help. I solved the problems by
>> installing GMT 4.1 instead of GMT 3.4. The grind, mapping, genex
>> modules are working fine under Linux OS.
>> Best regards
>> zehra
>>
>>
>> On 2/16/07, Bruce Worden <cbworden at usgs.gov> wrote:
>> Hmmm, that makes no sense to me. grdinfo can read the header, but
>> another program can't seek past it. Can you run grind and mapping
>> again, this time using the -verbose flag with each program (you'll
>> have to run them individually from the command line since "shake"
>> won't pass the verbose flag to its sub-programs -- use the -dryrun
>> flag to shake to get a listing of how it would run the sub-
>> programs, then just add -verbose to each in turn).
>>
>>
>> Also send me "mmi.grd" so I can see what is going on inside it.
>>
>>
>> Bruce
>>
>>
>>
>> On Feb 16, 2007, at 12:40 AM, zehra cagnan wrote:
>>
>>> Hi Bruce
>>>
>>> this is the output of grdinfo.
>>>
>>> thanks
>>>
>>> zehra
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: Title:
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: Command: /h ome/ozden/Desktop/SHAKEMAP/Kurulum/GMT/bin/
>>> nearneighbor - R-120.150000/-117.65000 0/33.729666/35.396333 -G/
>>> home/ozden/Desktop/SHAKEMAP/shake_home/bin/../data/9583 161/
>>> output/mmi.grd=1 - I0.0166666666666667/0.0166666666666667 -N1 -
>>> S0.01666666666 66667
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: Remark:
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: Normal node registration used
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: grdfile for mat # 1
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: x_min: -120 .15 x_max: - 117.65 x_inc: 0.0166667 units:
>>> user_x_unit nx: 151
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: y_min: 33.7 297 y_max: 35.3963 y_inc: 0.0166667 units:
>>> user_y_unit ny: 101
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: z_min: 1 z_ max: 4.46 units: user_z_unit
>>>
>>> /home/ozden/Desktop/SHAKEMAP/shake_home/data/9583161/output/
>>> mmi.grd: scale_facto r: 1 add_offset: 0
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On 2/16/07, Bruce Worden <cbworden at usgs.gov> wrote:
>>>
>>>
>>> What did the output of grdinfo look like?
>>>
>>>
>>> Bruce
>>>
>>> On Feb 15, 2007, at 2:13 PM, zehra cagnan wrote:
>>>
>>>> Hi Bruce,
>>>> I checked the macros and ShakeConfig.pm files. Architecture is
>>>> set to CISC.
>>>>
>>>> mmi.grd,pga.grd,pgv.grd they are each around 60 KB. On the other
>>>> hand I have approximately 10000 rows in the grid.xml file .
>>>> zehra
>>>>
>>>>
>>>>
>>>> On 2/15/07, Bruce Worden <cbworden at usgs.gov > wrote:
>>>> Hi Zehra,
>>>>
>>>>
>>>> On your second question -- there are some issues in getting to
>>>> the lists from outside the gps domain. We are going to move
>>>> shake-dev to a USGS site fairly soon, so you should have easier
>>>> access (as well as searchable archives). Have you tried this URL:
>>>>
>>>>
>>>> http://scsnmail.gps.caltech.edu/mailman/listinfo/shake-dev
>>>>
>>>>
>>>> I don't know if that will work better for you. I am inside the
>>>> domain, so I see it all just fine.
>>>>
>>>>
>>>> On your primary problem, your first error is:
>>>>
>>>>
>>>> GMT Fatal Error: Error seeking past header file /home/ozden/
>>>> Desktop/SHAKEMAP/shake_home/bin/../data/9583161/output/mmi.grd!
>>>>
>>>>
>>>> Which tells me that the .grd files are probably not being
>>>> created properly. Have a look at that file and see how big it
>>>> is. You can also use 'grdinfo' to examine the contents -- but
>>>> remember to append '=1' to the filename so grdinfo can read it
>>>> ( i.e., 'grdinfo mmi.grd=1'). If it gags, 'grind' isn't writing
>>>> the file correctly. If nx * ny * 4 isn't within a couple of
>>>> dozen bytes of the file size, there is some other problem.
>>>>
>>>>
>>>> Check your 'include/macros' file -- your 'ARCHITECTURE' macro
>>>> should be set to 'CISC'. If it isn't, you'll need to change it
>>>> and run "make all" in the shakemap home directory again. Then
>>>> make sure that the '$architecture' variable in lib/
>>>> ShakeConfig.pm is set to the proper thing. If it isn't, delete
>>>> the file and run 'make all' again.
>>>>
>>>>
>>>> Bruce
>>>>
>>>>
>>>>
>>>>
>>>> On Feb 15, 2007, at 5:11 AM, zehra cagnan wrote:
>>>>
>>>>> Hi Bruce
>>>>> I followed the shakemap manual and installed GMT version 3.4
>>>>> thinking that installing GMT version 4.1 would create problems.
>>>>> Thank you very much for the help you provided. I solved the
>>>>> problem below. I uncommented certain lines within mapping as
>>>>> they were applicable to verion 4.1 of GMT.
>>>>>
>>>>> The grind module seems to work with no problems but I am still
>>>>> experiencing problems with the mapping module. I am getting the
>>>>> error message below. Somehow mapping does not create the
>>>>> tmpmmi.grd file. I am seeing the mmi.grd file under the data/
>>>>> mapping directory but tmpmmi.grd file is not being created.
>>>>>
>>>>> Some of the files I need to complete regional configuration is
>>>>> not ready yet so I am using california data and trying to do a
>>>>> test run just to understand the program better.
>>>>> I commented out certain parts within shake.conf so that only
>>>>> grind,tag,mapping,genex modules will be called in the order I
>>>>> wrote above.
>>>>>
>>>>> My second question is regarding the shakemap list. I am trying
>>>>> to access the website of the list thinking that I might be able
>>>>> access the list archieves via there (may be some of the
>>>>> questions I am asking now were answered there before). I am
>>>>> using my email address as my user id and the password you sent
>>>>> me as my password but could not access the web site. is my user
>>>>> id something different than zcagnan at gmail.com
>>>>>
>>>>> thanks
>>>>> zehra
>>>>>
>>>>>
>>>>>
>>>>> grind: Type: pga; bias 0.815; absolute deviation was 0.019171
>>>>> is 0.017089
>>>>> grind: Type: pgv; bias 0.775; absolute deviation was 0.006004
>>>>> is 0.005444
>>>>> grind: 1 stations flagged out this iteration
>>>>> grind: Type: pga; bias 0.815; absolute deviation was 0.019171
>>>>> is 0.017089
>>>>> grind: Type: pgv; bias 0.775; absolute deviation was 0.006004
>>>>> is 0.005444
>>>>> grind: 0 stations flagged out this iteration
>>>>> GMT Fatal Error: Error seeking past header file /home/ozden/
>>>>> Desktop/SHAKEMAP/shake_home/bin/../data/9583161/output/mmi.grd!
>>>>> grdview: No such file or directory [/home/ozden/Desktop/
>>>>> SHAKEMAP/shake_home/bin/../data/9583161/mapping/tmp_mmi.grd]
>>>>> mapping: Error running command:
>>>>> /home/ozden/Desktop/SHAKEMAP/Kurulum/GMT/bin/grdview /home/
>>>>> ozden/Desktop/SHAKEMAP/shake_home/bin/../data/9583161/mapping/
>>>>> topo.grd -P - R-120.15/-117.65/33.729666/35.396333/10/3000 -
>>>>> JM6.5 -JZ1 -G/home/ozden/Desktop/SHAKEMAP/shake_home/bin/../
>>>>> data/9583161/mapping/tmp_mmi.grd -C/home/ozden/Desktop/SHAKEMAP/
>>>>> shake_home/bin/../lib/mapping/Ii.cpt -I/home/ozden/Desktop/
>>>>> SHAKEMAP/shake_home/bin/../data/9583161/mapping/inten.grd -Qi72
>>>>> -O -K|
>>>>>
>>>>> shake: Error in mapping: 256
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 2/14/07, Bruce Worden < cbworden at usgs.gov> wrote:
>>>>> Hi Zehra,
>>>>>
>>>>> What version of GMT are you using? There are problems with this
>>>>> version of ShakeMap and certain versions of GMT. The next
>>>>> version of
>>>>> ShakeMap (V3.2, which you can get a pre-release version of at
>>>>> the FTP
>>>>> site) should be more forgiving of GMT's idiosyncrasies.
>>>>>
>>>>> But in the version you have:
>>>>> Edit 'mapping' and search for the string "PLOT_DEGREE_FORMAT".
>>>>> There
>>>>> is some commentary there about fixing the first error, below.
>>>>>
>>>>> The topography error is odd -- can you verify that you have the
>>>>> topography files "cal_neva_9c[i].grd" (or whatever topo files you
>>>>> have configured in mapping.conf) and not just links? If they are
>>>>> links, you can get the actual files at the FTP site.
>>>>>
>>>>> Also, I am adding you to the shake-dev mailing list (you should
>>>>> get a
>>>>> confirmation email shortly -- make sure your spam filter
>>>>> doesn't get
>>>>> it) so you can ask your questions there, and maybe get more help.
>>>>>
>>>>> I hope this helps.
>>>>>
>>>>> Bruce
>>>>>
>>>>> On Feb 14, 2007, at 2:59 AM, zehra cagnan wrote:
>>>>>
>>>>> > Okey
>>>>> > I solved the previous problem. The binary form of the vsgrid.txt
>>>>> > that we were using was not generated on a linux machine. That
>>>>> was
>>>>> > the problem.
>>>>> > Now I am having problems with the GMT.
>>>>> >
>>>>> > The error message that i am getting is below. Any help you can
>>>>> > provide?
>>>>> >
>>>>> > grind: Type: pga; bias 0.815; absolute deviation was 0.019171 is
>>>>> > 0.017089
>>>>> > grind: Type: pgv; bias 0.775; absolute deviation was 0.006004 is
>>>>> > 0.005444
>>>>> > grind: 1 stations flagged out this iteration
>>>>> > grind: Type: pga; bias 0.815; absolute deviation was 0.019171 is
>>>>> > 0.017089
>>>>> > grind: Type: pgv; bias 0.775; absolute deviation was 0.006004 is
>>>>> > 0.005444
>>>>> > grind: 0 stations flagged out this iteration
>>>>> > gmtset: GMT SYNTAX ERROR in GMT_setparameter: Unrecognized
>>>>> keyword
>>>>> > PLOT_DEGREE_FORMAT
>>>>> > gmtset: GMT SYNTAX ERROR: PLOT_DEGREE_FORMAT given illegal
>>>>> value (D)!
>>>>> > gmtset: 1 conversion errors
>>>>> > mapping: WARNING: no topography data found; will continue
>>>>> without
>>>>> > topo.
>>>>> > GMT Fatal Error: Error seeking past header file /home/ozden/
>>>>> Desktop/
>>>>> > SHAKEMAP/shake_home/bin/../data/9583161/output/mmi.grd!
>>>>> > mapping: WARNING: can't find topo data: making mmi without
>>>>> topography
>>>>> > grdimage: No such file or directory [/home/ozden/Desktop/
>>>>> SHAKEMAP/
>>>>> > shake_home/bin/../data/9583161/mapping/tmp_mmi.grd]
>>>>> > mapping: Error running command:
>>>>> > /home/ozden/Desktop/SHAKEMAP/Kurulum/GMT/bin/grdimage /home/
>>>>> ozden/
>>>>> > Desktop/SHAKEMAP/shake_home/bin/../data/9583161/mapping/
>>>>> tmp_mmi.grd
>>>>> > -P -R-120.15/-117.65/33.729666/35.396333/10/3000 - JM6.5 -C/
>>>>> home/
>>>>> > ozden/Desktop/SHAKEMAP/shake_home/bin/../lib/mapping/Ii.cpt -
>>>>> O -K|
>>>>> >
>>>>> >
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Shake-dev mailing list
>>>> Shake-dev at scsnmail.gps.caltech.edu
>>>> http://unix1.gps.caltech.edu/mailman/listinfo/shake-dev
>>>
>>>
>>>
>>> _______________________________________________
>>> Shake-dev mailing list
>>> Shake-dev at scsnmail.gps.caltech.edu
>>> http://unix1.gps.caltech.edu/mailman/listinfo/shake-dev
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geohazards.usgs.gov/pipermail/shake-dev/attachments/20070308/d7853f5d/attachment-0002.html
More information about the Shake-dev
mailing list